Text Summarization Adaptive Models for Semantic Relevance Information: A Survey

Abstract

Text Summarization has long been a prominent field of research in the Artificial Intelligence. Text summarising is the method to extract the essential ideas and meaning from a given text and trying to turn them into a summary. Automatic summarization has evolved into a crucial technique for quickly and efficiently finding important information in large amounts of text. It is really challenging to summarize large amount of data but by utilising various Natural Language approaches, it can be accomplished with pre-trained models, a review of adaptive models for summarization is presented in this study such as ‘DA-PN + Cover + MLO’ model, Auto-Encoder (AE), Variational Auto Encoder (VAE), Extreme Learning Machine Auto-Encoder (ELMAE) and Text Rank algorithm. The main objectives of this study are to auto-summarize the text given by user and to estimate idea importance and choose the lines that will be contained in the conclusion that are the most important.